Open Source Controller with internal THC

Hi thefabricator03, I want you to know that I respect your opinion. From your point of view, maybe things look like this. But I don't think you're right. In fact, you barely even understood what I meant, right?
According to what you wrote, you look like a very above average Linux user. Very good of you, it would be to be broad-minded and to be able to easily get into someone else's shoes. Then of course you would know the statistics. on the question (you should be interested in this, right). Does anyone actually care how many people can compile themselves for a particular computer - LinuxCNC?
Then of course you wouldn't write that anyone could install the unfinished one left somewhere in the years of Linux (we won't argue since when the default screen and installer was made and it hasn't been changed until today), collecting it from different places on the Internet and supplement it with various additives to make it look like a complete CNC system.
On the other hand, if you try to understand me, you will see that while we can use a cheap resource from the Chinese, (the small pc I mentioned) it is good to create a LinuxCNC for them to increase the number of people who use it. This will only bring good to the community and its users. In the end, if it is developed by a specific programmer, we can help him sell it!
Only in this way, a good idea can be improved and acquire the qualities of a real CNC system, which, unfortunately, LinuxCNC is not yet. I will not go into details, because I am sure that I will need pages of text to explain my position.
I will only add that, in my opinion, until the real time is moved from the computer, the PC based CNCs will not move forward in their development. If this happens, the PC will stop being a problem and everything will be processed on the cheap and super accurate FPGA it will be ok.
Again, I want to point out that I am writing this with good feelings, despite the strange English!

P.P It is a pity that your opinion is this tommylight
P.P In fact, I originally installed LinuxCNC 2.7 on a MITAC 210 (because it has a DB25 printer port).This is how my metal mill works for several months.
Well, a great movie, until I understand the way of thinking of the poor man who created the HOME procedure ...:laugh:
Then I delved into installing thousands of nonsense created by different people to fix unfinished things, and finally I read so many unnecessary pages ....
Now I ordered two T95 Intel Cherry Trail Z8350 and I plan to try again - this time with Mesa 7I96 + THCAD

I don't think thefabricator knew much about linux when he started with linuxcnc at all. What he knows he's learnt from necessity becasue the knowledge was required to use his chosen tool.

I can't see that Linuxcnc will ever move the CNC controller to a FGPA card. The Linux world is slowly moving towards joining linuxcnc with a real time kernel as there are now patches that can install the required PREEMPT_RT kernel on Debian and other distributions. Some of us had no choice but to compile the kernel from scratch.

This is driven by some major players (BMW is one) who require a real time OS. Ultimately, I believe that the Real time kernel will become the default for Linux across the board. Thats certainly my information. The PC is a very low cost component on the average CNC build. With the advent of compatible single board computers like the Raspberry PI and Odroid H2+, there is no need to throw out LInuxcnc's core in favour of an emasculated FGPA or other microprocessor. Linuxcnc supports remote network access. Why not embed an odroid in the machine and run it via a network connection? I its increasingly more likely that the linuxcnc model will gain greater recognition moving forward as the PREMEPT_RT kernel gains mainstream recognition becasue a PC has infinitely more resourced than a microprocessor.

Hello rodw, you are somewhat right. But my opinion is that if the place of the computer is in the dashboard of the drivers, you will have a problem with the monitor (for larger machines, such as plasmas 6x2 meters, for example) - we have already done that, it is not good!
On the other hand, why should you get to pseudo real time if something else can be done - an easier solution. Which in the long run will allow LinuxCNC to no longer pour out of the computer. There are so many more important things to do , and we are still discussing real time. I think that the computer part should deal - with code analysis, traffic planning and creating a series of simple actions to be submitted through the FPGA buffer for execution. On board the FPGA there is real time and everything will happen just in time and with smooth and precise steps. FPGA will return "1-OK, 2-Ok, 3-Ok ...." and so LinuxCNC will refresh the toolpath as well as the DRO.

What kind of problem you have with monitor?

what i have been able to understand @stivemaster13 wants Commericial like embedded hardware with software from open source movement. Moreover, he wants to build and sell those solutions as propreitory(hiding linuxcnc); Like, the aussie masso cnc controllers.
I know this because I have been at that place.

Pay some programmers and pcb designers, you can get your thing working, you can even use linuxcnc and modify it as tormach does.


Thing is Commercial are commercial.
Making open software limited to a hardware is not a right thing to do.

10000$ hardware+software
or
200$ your own hardware + software
Choice is yours.

Dont blame linuxcnc for not making your dreams controllers easy enough.
Cheers,
Pinder

pinder wrote: what i have been able to understand @stivemaster13 wants Commericial like embedded hardware with software from open source movement. Moreover, he wants to build and sell those solutions as propreitory(hiding linuxcnc); Like, the aussie masso cnc controllers.
I know this because I have been at that place.
Pinder

Is Masso Linuxcnc under the hood?